Christopher Wills, a professor of biology from UCSD, and an ID critic, acknowledges that the ID hypothesis “is in principle testable.” What is most striking about Wills’s claims is that they are not indebted to the designer-centric approach. This demonstrates that, in the end, the designer-centric approach is a matter of taste and convenience, not necessity.

Nevertheless, one of Wills’s test is embedded with thorny issues.

One of Wills’s suggested lines of evidence is as follows:

Or, in the case of evolution, one could search for sudden discontinuities in the history of life, in which a new structure or function has arisen without any previous history and no relationship to structures or functions in other related organisms. (Such new structures have not yet been found, by the way.)

Of course, as many have noted, this sounds more like evidence against evolution than evidence for intelligent design. Yet as I have shown, most critics of ID demand some type of anti-evolutionary evidence as evidence for design. For example, if I employ teleological logic to infer a proofreading ability for RNA polymerase, the main complaint is that I have failed to rule out evolution and its ability to generate this property. Or, if we begin to consider the hypothesis of front-loaded evolution (FLE), the common complaint I have encountered is that FLE is supposed to uncover something that evolution cannot explain. Thus, regardless of whether Wills’s test does in fact translate as evidence for ID, the fact remains that many people, on both sides of the aisle, think it does. And there is reason to think they are correct. But that’s another topic.

However, we have to be very careful with Wills’s test, as it constrains the designer to designing only unique things. In fact, some have commented that the designer would be “lazy” if similar designs were used in different contexts. This constraint actually works to skew the analysis.

When Wills says we need to find “a new structure or function has arisen without any previous history,” we don’t really have a time machine to determine whether something has a previous history. Instead, what we do is infer such a history based on relationships to structures or functions in other related organisms. Thus, the only way to rule out a previous history is to find something that has no functional or structural relationship with other organisms. But just how deep must the uniqueness run?

Consider a intelligent agent designing a gram-negative bacterium (like E. coli) and a gram-positive bacterium (like B. subtilis). We could point out that the architecture of the cell walls are different and B. subtilis can form spores while E. coli can’t. If we compared only these two, the additional membrane of E. coli and the spores of B.subtilis would appear to satisfy Wills’s test. But if we look closer at these two species, we’d find the cell walls are composed of the same material. Is that a signal of common descent or a common design strategy? Furthermore, we’d find loads of other similarities in terms of cell division, protein composition, metabolism, DNA processing, transcription, and translation. What’s more, the amino acids used are the same, the nucleotides are the same, and the genetic code is the same. The two bacteria would fails Wills’s test.

To pass his test, E. coli and B. subtilis would have be different in every way. They’d have to use different amino acids, different nucleotides, a different genetic code, and a different metabolism. Anything that has a functional or structural relationship with other organisms would signal common descent. In fact, if the only thing shared by E. coli and B. subtilis was a common genetic code and a common use of the same building blocks, this would be interpreted as evidence that the common ancestor was very ancient, existing prior to the evolution of the different machinery.

If E. coli and B. subtilis passed this test, I think it rather obvious what would happen.

First, many non-teleologists would argue that this is evidence that the prebiotic soup had plenty of degrees of freedom from which to spawn all kinds of life forms and thus strongly supports Gould’s view about replaying life’s tape.

Secondly, some critics would argue that, even from a design perspective, these differences are evidence of multiple designers.

So by insisting that a designer is constrained to designing only unique things, the situation is rigged to channel our explanation of similarities toward common descent. If discontinuties are found, they are channeled toward a belief that biological components can be easily spawned by chance or an insistence on multiple designers.

The problem? Our experience with design teaches us that there is no such constraint; designers are not in any way obligated to re-invent the wheel every time they sit down at the drawing board. At some point, design theorists are going to have to grapple with Common Design in a truly investigative manner. I’ve taken some first baby steps in this direction here.